# How does PyTorch pick cuda versions?

**URL:** <https://discuss.pytorch.org/t/how-does-pytorch-pick-cuda-versions/137941>\
**Category:** Uncategorized\
**Created:** [November 27, 2021, 1:06am UTC](https://discuss.pytorch.org/t/how-does-pytorch-pick-cuda-versions/137941 "2021-11-27T01:06:46Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![cjenkins5614](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/cjenkins5614/32/41284_2.png) [@cjenkins5614](https://discuss.pytorch.org/u/cjenkins5614)\
**Post date:** [November 27, 2021, 1:06am UTC](https://discuss.pytorch.org/t/how-does-pytorch-pick-cuda-versions/137941/1 "2021-11-27T01:06:46Z")

</div>

I always use the standard download selection panel to match cuda versions ([https://pytorch.org/](https://pytorch.org/)):

```auto
pip3 install torch==1.8.2+cu102 torchvision==0.9.2+cu102 torchaudio==0.8.2 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html

```

My two questions:

1. Does PyTorch look at strictly `/usr/local/cuda`'s linkage and decide what directory to dig into? If I have `cuda` linked to `cuda-10.1`, but have `cuda-10.2` around, what would `torch==x.y.z+cu102` pick?

```auto
--> ls -hdlt /usr/local/cuda*
drwxr-xr-x 20 root root 4.0K Oct 9 13:00 /usr/local/cuda-10.2/
lrwxrwxrwx 1 root root 20 Jul 22 13:37 /usr/local/cuda -> /usr/local/cuda-10.1/
drwxr-xr-x 25 root root 4.0K Jun 23 21:23 /usr/local/cuda-11.0/
drwxr-xr-x 19 root root 4.0K Apr 22 2021 /usr/local/cuda-10.1/
drwxr-xr-x 19 root root 4.0K Apr 22 2021 /usr/local/cuda-11.1/

```

1. I’ve found that even if there were only one `cuda-10.1` in `/usr/local`, and `torch==x.y.z+cu102` would still work, most of the time. When do I expect it not to work, and what’s the caveat of having mismatches, minor or major?

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [November 27, 2021, 1:43am UTC](https://discuss.pytorch.org/t/how-does-pytorch-pick-cuda-versions/137941/2 "2021-11-27T01:43:34Z")

</div>

Your local CUDA toolkit won’t be used unless you build PyTorch from source or a custom CUDA extension, since the pip wheels and conda binaries use their own CUDA runtime.  
You would thus only need to install a valid NVIDIA driver and can use the binaries directly.

---

<div class="post-metadata">

**Author:** ![cjenkins5614](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/cjenkins5614/32/41284_2.png) [@cjenkins5614](https://discuss.pytorch.org/u/cjenkins5614)\
**Post date:** [November 27, 2021, 2:14am UTC](https://discuss.pytorch.org/t/how-does-pytorch-pick-cuda-versions/137941/3 "2021-11-27T02:14:33Z")

</div>

That explains… So the choice really depends on whether the driver is compatible with the pytorch cuda version to be installed.